Re: Passive Wave Primer

[James Jun <james.jun () towardex com>]

On Tue, Oct 13, 2020 at 12:27:44PM +0000, Rod Beck wrote:
> Dear Network Gurus,
>
> Looking for a tutorial on passive waves. How it works. Pros and cons. .

Essentially, you're providing a channel off of your DWDM filters for someone else to pass light.

Commonly in the market, a "wavelength" product generally isn't a true wavelength, especially on long-haul segments.
The 'wavelength' market really is an evolution of the old SONET market in some ways -- carriers will typically light a 
channel (either in fixed grid filter or flex grid) and that single channel is usually an X-gigabaud (e.g. 35-95Gbd) 
that uses coherent modulation on line side for say 200-800Gbps and multiplexing for tributary channels (such as TDM) on 
client side ports to break away a 100GE circuit for the customer end-user.

As far as technicalities are concerned, most 'wavelength' products that behave as described above, ought to be called 
"dedicated circuits" or "circuit-switched transport" if we're anal about its operating principles.

As for 'true' wavelength service, that brings us to your question:

When you're talking about passive wave or 'alien wave', what you're doing is you're providing a wavelength frequency 
assignment on your photonic filter system (a channel on your 100 Ghz fixed grid DWDM filter, or bandwidth assignment 
window on your flex grid ROADM) to the customer, which would typically be another network provider, or a very clued 
enterprise customer that wants to run his own optical transport but can't justify the economics of full dark fiber over 
the said span, and doesn't need more than <=95Gbd max of modulation bandwidth.

The customer would pass traffic similarly to how you yourself would light a channel, installing a coherent transponder 
for 200-800Gbps wave facing the line side, and breaking it out to Nx100GE for end-user traffic.

James